JuliaDB compatibility issue

Hi!

I’m trying to use JuliaDB. When I try to add it, I get the following incompatibility message:

(@v1.7) pkg> add JuliaDB
   Resolving package versions...
ERROR: Unsatisfiable requirements detected for package MLUtils [f1d291b0]:
 MLUtils [f1d291b0] log:
 ├─possible versions are: 0.1.0-0.2.6 or uninstalled
 ├─restricted to versions * by an explicit requirement, leaving only versions 0.1.0-0.2.6
 └─restricted by compatibility requirements with StatsBase [2913bbd2] to versions: uninstalled — no versions left
   └─StatsBase [2913bbd2] log:
     ├─possible versions are: 0.24.0-0.33.16 or uninstalled
     ├─restricted to versions * by an explicit requirement, leaving only versions 0.24.0-0.33.16
     └─restricted by compatibility requirements with JuliaDB [a93385a2] to versions: 0.24.0-0.32.2
       └─JuliaDB [a93385a2] log:
         ├─possible versions are: 0.9.0-0.13.1 or uninstalled
         └─restricted to versions * by an explicit requirement, leaving only versions 0.9.0-0.13.1

I reckon that StatsBase and JuliaDB are incompatible. Have I understood the message correctly?
Is there a solution to this problem?

Thanks a lot,
Inbar

You haven’t understood it quite correctly. Let me post a screenshot - normally discouraged, but I’ll do it for the colours - of what happens when I try to add JuliaDB and MLUtils in a new environment [NB: don’t just dump all your projects into the main environment, work with project specific environments]:

hopefully this makes it clearer:

  • both MLUtils and JuliaDB depend on StatsBase
  • MLUtils requires StatsBase >= 0.33
  • JuliaDB requires StatsBase <= 0.32.2

The best way around this would be to bump the compat bounds of JuliaDB, but unfortunately it hasn’t been updated in a while.

A hack would be to dev JuliaDB locally and just relax the compat to allow StatsBase 0.33 to see if it works - if it does you could upstream that change (and maybe become a new JuliaDB maintainer!)

2 Likes

Given that JuliaDB.jl is not maintained currently could you say what is your usecase as maybe some alternatives could be recommended instead?

I try to teach myself statistics and (postgre)SQL at the same time, so instead of storing my simulation samples “manually” in a text file I wanted to save them to a database and pull them back later for further analysis.

You don’t need JuliaDB for that, see:

https://juliadatabases.org/

In particular using GitHub - invenia/LibPQ.jl: A Julia wrapper for libpq in combination with GitHub - JuliaData/DataFrames.jl: In-memory tabular data in Julia should be sufficient for most learning purposes.

2 Likes